0bps

Strategy vault / Trend following

Configurable Dual direction Supertrend Strategy

This strategy is named "Configurable Dual-direction Supertrend Strategy". It uses the Supertrend trailing stop mechanism to identify price trends, and allows separate parameter configuration for long and short…

Author
ChaoZhang
Complexity
Intermediate · 165 lines
Parameters
32
Backtest bar
3h · intraday
Tested on
BTC/USDT · futures

This strategy is named "Configurable Dual-direction Supertrend Strategy". It uses the Supertrend trailing stop mechanism to identify price trends, and allows separate parameter configuration for long and short trades, enabling precise trend following.

The Supertrend calculation is: using ATR multiplied by a coefficient to build price channels. The upper band is the long stop loss and the lower band is the short stop loss. Price breaking the channel generates trade signals.

The innovation is the independent parameter configuration for long and short:

  1. Supertrend parameters like ATR period and coefficient can be set separately.

  2. Maximum holding period can also be configured independently to adjust profit targets.

  3. Stop loss methods (fixed percentage or ATR trailing) can also be set differently.

This allows only-long, only-short or dual-direction trading to better fit specific market conditions.

The advantages are the intuitive Supertrend mechanism and abundant configurable combinations. But Supertrend alone is prone to breaches and needs confirmation. Parameter optimization is also crucial.

In summary, the configurable dual Supertrend strategy improves trend trading precision, while keeping the core idea simple for practical application.

Parameters

NameDefaultWhat it controls
v_input_1false════════ Test Period ═══════
v_input_22017Backtest Start Year
v_input_3trueBacktest Start Month
v_input_4trueBacktest Start Day
v_input_52019Backtest Stop Year
v_input_612Backtest Stop Month
v_input_731Backtest Stop Day
v_input_8false═════ Super Trend L ═════
v_input_92ATR Period
v_input_101.5ATR Multiplier
v_input_11false═════ Super Trend S ═════
v_input_123ATR Period
v_input_131.3ATR Multiplier
v_input_14false═════ Rate of Change L ═════
v_input_1530ROC Length
v_input_166ROC % Change
v_input_17false═════ Rate of Change S ═════
v_input_1876ROC Length
v_input_196ROC % Change
v_input_20false═══════ Stop Loss L ══════
v_input_210Stop Loss Type: Fixed | ATR Derived
v_input_226Fixed Stop Loss %
v_input_2320ATR Stop Period
v_input_241.5ATR Stop Multiplier
v_input_25false═══════ Stop Loss S ══════
v_input_260Stop Loss Type: Fixed | ATR Derived
v_input_276Fixed Stop Loss %
v_input_2820ATR Stop Period
v_input_291.5ATR Stop Multiplier
v_input_30false══════ Longs or Shorts ═════
v_input_31trueUse Longs
v_input_32trueUse Shorts

Source Pine Script

/*backtest
start: 2023-01-01 00:00:00
end: 2023-09-12 00:00:00
period: 3h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
args: [["v_input_8",true],["v_input_11",true]]
*/

//@version=4
strategy("Super Trend Daily 2.0 BF ?", overlay=true, precision=2, initial_capital=10000, default_qty_type=strategy.percent_of_equity, default_qty_value=100, commission_type=strategy.commission.percent, commission_value=0.075)

/////////////// Time Frame ///////////////
_0 = input(false,  "════════ Test Period ═══════")
testStartYear = input(2017, "Backtest Start Year") 
testStartMonth = input(1, "Backtest Start Month")
testStartDay = input(1, "Backtest Start Day")
testPeriodStart = timestamp(testStartYear,testStartMonth,testStartDay, 0, 0)

testStopYear = input(2019, "Backtest Stop Year")
testStopMonth = input(12, "Backtest Stop Month")
testStopDay = input(31, "Backtest Stop Day")
testPeriodStop = timestamp(testStopYear,testStopMonth,testStopDay, 0, 0)

testPeriod() => true

///////////// Super Trend Long /////////////
_1 = input(false,  "═════ Super Trend L ═════")
lengthl = input(title="ATR Period", type=input.integer, defval=2)
multl = input(title="ATR Multiplier", type=input.float, step=0.1, defval=1.5)

atrl = multl * atr(lengthl)

longStopl = hl2 - atrl
longStopPrevl = nz(longStopl[1], longStopl)
longStopl :=  close[1] > longStopPrevl ? max(longStopl, longStopPrevl) : longStopl

shortStopl = hl2 + atrl
shortStopPrevl = nz(shortStopl[1], shortStopl)
shortStopl := close[1] < shortStopPrevl ? min(shortStopl, shortStopPrevl) : shortStopl

dirl = 1
dirl := nz(dirl[1], dirl)
dirl := dirl == -1 and close > shortStopPrevl ? 1 : dirl == 1 and close < longStopPrevl ? -1 : dirl

///////////// Super Trend Short /////////////
_2 = input(false,  "═════ Super Trend S ═════")
lengths = input(title="ATR Period", type=input.integer, defval=3)
mults = input(title="ATR Multiplier", type=input.float, step=0.1, defval=1.3)

atrs = mults * atr(lengths)

longStops = hl2 - atrs
longStopPrevs = nz(longStops[1], longStops)
longStops :=  close[1] > longStopPrevs ? max(longStops, longStopPrevs) : longStops

shortStops = hl2 + atrs
shortStopPrevs = nz(shortStops[1], shortStops)
shortStops := close[1] < shortStopPrevs ? min(shortStops, shortStopPrevs) : shortStops

dirs = 1
dirs := nz(dirs[1], dirs)
dirs := dirs == -1 and close > shortStopPrevs ? 1 : dirs == 1 and close < longStopPrevs ? -1 : dirs

///////////// Rate Of Change Long ///////////// 
_3 = input(false,  "═════ Rate of Change L ═════")
sourcel = close
roclengthl = input(30, "ROC Length",  minval=1)
pcntChangel = input(6, "ROC % Change", minval=1)
rocl = 100 * (sourcel - sourcel[roclengthl]) / sourcel[roclengthl]
emarocl = ema(rocl, roclengthl / 2)
isMovingl() => emarocl > (pcntChangel / 2) or emarocl < (0 - (pcntChangel / 2))

///////////// Rate Of Change Short ///////////// 
_4 = input(false,  "═════ Rate of Change S ═════")
sources = close
roclengths = input(76, "ROC Length",  minval=1)
pcntChanges = input(6, "ROC % Change", minval=1)
rocs = 100 * (sources - sources[roclengths]) / sources[roclengths]
emarocs = ema(rocs, roclengths / 2)
isMovings() => emarocs > (pcntChanges / 2) or emarocs < (0 - (pcntChanges / 2))

/////////////// Strategy /////////////// 
long = dirl == 1 and dirl[1] == -1 and isMovingl()
short = dirs == -1 and dirs[1] == 1 and isMovings()

last_long = 0.0
last_short = 0.0
last_long := long ? time : nz(last_long[1])
last_short := short ? time : nz(last_short[1])

long_signal = crossover(last_long, last_short)
short_signal = crossover(last_short, last_long)

last_open_long_signal = 0.0
last_open_short_signal = 0.0
last_open_long_signal := long_signal ? open : nz(last_open_long_signal[1])
last_open_short_signal := short_signal ? open : nz(last_open_short_signal[1])

last_long_signal = 0.0
last_short_signal = 0.0
last_long_signal := long_signal ? time : nz(last_long_signal[1])
last_short_signal := short_signal ? time : nz(last_short_signal[1])

in_long_signal = last_long_signal > last_short_signal
in_short_signal = last_short_signal > last_long_signal

last_high = 0.0
last_low = 0.0
last_high := not in_long_signal ? na : in_long_signal and (na(last_high[1]) or high > nz(last_high[1])) ? high : nz(last_high[1])
last_low := not in_short_signal ? na : in_short_signal and (na(last_low[1]) or low < nz(last_low[1])) ? low : nz(last_low[1])

since_longEntry = barssince(last_open_long_signal != last_open_long_signal[1]) 
since_shortEntry = barssince(last_open_short_signal != last_open_short_signal[1]) 

/////////////// Stop Losses Long ///////////////
_5 = input(false,  "═══════ Stop Loss L ══════")
SL_typel = input("Fixed", options=["Fixed", "ATR Derived"], title="Stop Loss Type")
sl_inpl = input(6.0, title='Fixed Stop Loss %') / 100
atrLkbl = input(20, minval=1, title='ATR Stop Period')
atrMultl = input(1.5, step=0.25, title='ATR Stop Multiplier') 
atr1l = atr(atrLkbl)

longStop1l = 0.0
longStop1l :=  short_signal ? na : long_signal ? close - (atr1l * atrMultl) : longStop1l[1]

slLongl = in_long_signal ? strategy.position_avg_price * (1 - sl_inpl) : na
long_sll = in_long_signal ? slLongl : na

/////////////// Stop Losses Short ///////////////
_6 = input(false,  "═══════ Stop Loss S ══════")
SL_types = input("Fixed", options=["Fixed", "ATR Derived"], title="Stop Loss Type")
sl_inps = input(6.0, title='Fixed Stop Loss %') / 100
atrLkbs = input(20, minval=1, title='ATR Stop Period')
atrMults = input(1.5, step=0.25, title='ATR Stop Multiplier') 
atr1s = atr(atrLkbs)

shortStop1s = 0.0
shortStop1s := long_signal ? na : short_signal ? close + (atr1s * atrMults) : shortStop1s[1]

slShorts = strategy.position_avg_price * (1 + sl_inps)
short_sls = in_short_signal ? slShorts : na

_7 = input(false,  "══════ Longs or Shorts ═════")
useLongs = input(true, title="Use Longs")
useShorts = input(true, title="Use Shorts")

/////////////// Execution ///////////////
if testPeriod()
    if useLongs
        strategy.entry("L", strategy.long, when=long)
        strategy.exit("L SL", "L", stop = SL_typel == "Fixed" ? long_sll : longStop1l, when=since_longEntry > 0)
    if useShorts
        strategy.exit("S SL", "S", stop = SL_types == "Fixed" ? short_sls : shortStop1s, when=since_shortEntry > 0)
        strategy.entry("S", strategy.short, when=short)
    if not useShorts
        strategy.close("L", when=short)
    if not useLongs
        strategy.close("S", when=long)

/////////////// Plotting /////////////// 
bgcolor(long_signal ? color.lime : short_signal ? color.red : na, transp=30)
bgcolor(not isMovings() ? color.white : not isMovingl() ? color.aqua : na)
plot(strategy.position_size <= 0 ? na : SL_typel == "Fixed" ? long_sll : longStop1l, title="Long Stop Loss", color=color.yellow, style=plot.style_circles, linewidth=2)
plot(strategy.position_size >= 0 ? na : SL_types == "Fixed" ? short_sls : shortStop1s, title="Short Stop Loss", color=color.orange, style=plot.style_circles, linewidth=2)

Published by ChaoZhang on the FMZ strategy square · last modified 2023-09-13. Collected in the Quant Trading Vault. Unvalidated research: re-test with your own data and costs before trusting it.

Build this with AgenKit

A spec is not a trading system. Paste this into your coding agent and AgenKit runs a specialist team — architecture, test-first build, risk gates, review — with paper trading on by default.

in Claude Code, Cursor, Codex & 4 more
/agenkit build a production-grade trading bot with a backtest engine, risk controls and paper trading by default from strategies/Configurable-Dual-direction-Supertrend-Strategy.md
  1. npx agenkit install engineering-kit in your project
  2. Paste the command into your agent and approve each phase
  3. Get tests, backtest, risk controls and a paper-trading switch

Understand it first

Before you run this, read the lesson behind the idea and how to backtest it honestly. Specs like this one are the survivors of an unknown number of trials; deflate their results accordingly.

Related specs

esc

Try , , or .

↑↓ move↵ open